2016 Postmortem
Related: About this forumMonmouth's methodology for Iowa poll pretty much says it all.
I was going to post this in my other thread, but that grew so fast this would be lost in the shuffle!
http://www.monmouth.edu/assets/0/32212254770/32212254991/32212254992/32212254994/32212254995/30064771087/c83c0429-6c12-47a7-ac86-9bd1beab9bfb.pdf
The Monmouth University Poll was sponsored and conducted by the Monmouth University Polling Institute from October 22 to 25, 2015 with a statewide random sample of 400 Iowa voters drawn from a list of registered Democratic voters who voted in at least one of the last two state primary elections and indicate they are likely to attend the Democratic presidential caucuses in February 2016. This includes 300 contacted by a live interviewer on a landline telephone and 100 contacted by a live interviewer on a cell phone, in English.
>400 is pretty small sample size
>already registered, already voted in at least one of last two primaries
>75% landline telephones
7% 18-34
17% 35-49
37% 50-64
39% 65+
>very heavily skewed toward older voters
ChairmanAgnostic
(28,017 posts)figures.
MohRokTah
(15,429 posts)Recced this thread as it is another unskewing thread
workinclasszero
(28,270 posts)Oh my...
NurseJackie
(42,862 posts)Personally, I think individual polls are a poor indicator of exactly what's going on any given day, and in response to any particular news event. But it's a good indicator of overall trends when all the polls are taken as a whole. It's very easy to nitpick any individual pollster or their methods, but it's difficult to argue-away obvious trends.
This poll is a piece of a puzzle. Not the entire puzzle, just a piece. When put together with other polls, we'll be able to come up with a clearer picture.
Capt. Obvious
(9,002 posts)are you saying that Bernie is nominated for Sexiest Man Of The Year?
magical thyme
(14,881 posts)with a smallish sample size that left out all new voters, not yet registered voters, voters who didn't attend one of the last 2 caucuses, and included mostly older voters.
NurseJackie
(42,862 posts)It's difficult for me to imagine that Monmouth would be so respected if they were as biased or incompetent as you're suggesting.
Response to magical thyme (Original post)
Name removed Message auto-removed
RoccoR5955
(12,471 posts)saying that this may not be good either. I never trust those polls. They are often designed to get the answer that one wants.
NCTraveler
(30,481 posts)What was shocking to me was the number of Sanders supporters who saw Hillary as their second choice instead of O'Malley. There isn't this "anybody but Hillary" sentiment in the real world that you find on du. Thanks for sharing. Lots of great information.
HerbChestnut
(3,649 posts)"Adults in the room"? Sheesh...
NCTraveler
(30,481 posts)I think politically it is a completely accurate statement. Look at the numbers. It clearly shows who the "political adults" are supporting overwhelmingly. With the weighting they used, I would make the case it is great for her no matter how it is broken down.
Capt. Obvious
(9,002 posts)and tell you he did it.
He may even post a
MohRokTah
(15,429 posts)that even 100 s is understating the hilarity.
DCBob
(24,689 posts)"Final sample is weighted for age and gender based on state registration list information on the pool of voters who
participate in primary elections."
Page 4:
http://www.monmouth.edu/assets/0/32212254770/32212254991/32212254992/32212254994/32212254995/30064771087/c83c0429-6c12-47a7-ac86-9bd1beab9bfb.pdf
Godhumor
(6,437 posts)The Iowa Democratic voting population is small, so a sample of 400 is not small relative to the population being polled.
randome
(34,845 posts)Since most people, as you've implied, support him. Why would it matter what a flawed poll indicates?
[hr][font color="blue"][center]Stop looking for heroes. BE one.[/center][/font][hr]
Mass
(27,315 posts)samples are so small, but they are. Kind of funny however to see the pundits argue on the meaning of a 1% difference in a poll whose MOE is 5 or 6 % (in the GOP primary, of course).
The two other conditions are also pretty much generic these days. They may be wrong, but they are there whether the poll shows HRC or Sanders ahead.
Which is more interesting is the age difference. However, this goes to a question I cannot find anybody able to answer. Who are the people who answer polls these days? It is actually even more problematic in the GOP primary. but what is the validity of polls where only 1 out 10 people accept to answer? How much does that skew the results (one way of the other)?
So, I would not spend too much time on polls, by which I mean whoever they favor. I see a lot of people argue that polls are wrong when they do not go their way. They should also argue that when they go their way.
winter is coming
(11,785 posts)If so, what were the results? If not, why not?
magical thyme
(14,881 posts)You'd have to ask them why they haven't polled in Iowa before now.
The nearest comparison is the Loras College poll which which got similar numbers just before the hearing, and which also screens out younger and new voters who have not yet registered, including only people who voted in 2012 and 2014.